DESCRIPTION:This year, nearly 1.4 million Americans will be diagnosed with cancer. Most will survive 5 years and be considered "cured." However, cancer survivors are at increased risk for developing other cancers, diabetes and/or cardiovascular disease. Reasons for increased risk across conditions may be genetic, treatment-related or because common lifestyle factors increase risk for more than one disease. Thus, if cancer survivors adopt healthier lifestyles, they also may reduce their risk for secondary cancers and other chronic disorders. Our preliminary study data on 988 prostate and breast cancer patients suggest that cancer survivors are exceptionally receptive toward improving their diet and physical activity behaviors, and that the cancer diagnosis may render a" teachable moment." Few efforts have capitalized on this apparent opportunity to promote behavioral change within this high risk population - one reason may be the barrier of distance (travel), which cancer survivors list as a leading reason for non-participation in health promotion programs. Fortunately, there are several ways to overcome this barrier. An experienced team of investigators in the fields of nutrition, exercise, behavior, aging, biostatistics & medical informatics at Duke Univ. Med. Ctr. proposes to test the efficacy of a tailored correspondence course in changing both diet & physical activity behaviors. We will recruit 530 early stage prostate & breast cancer patients in 25 NC counties & randomize them to: 1) a group that receives a 10 mth. tailored correspondence course that promotes increased physical activity, a low fat diet, and increased fruit & vegetable intake; or 2) a usual care control group that receives information in unrelated areas. Intervention efficacy will be gauged by the proportion of participants in the intervention arm who adopt two or more goal behaviors compared to control group participants. The purpose of this study is to determine: 1) the relative short (post-intervention) & long-term (1 yr follow-up) efficacy of this personalized, computer-generated intervention; 2) the effects of the intervention on other endpoints (quality of life, perceived health, etc); & 3) factors, such as gender & race that may interact with the intervention in predicting program efficacy. If effective, this program could be modified, easily disseminated & used in a variety of health care settings, thus having broad public health impact.